AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

ResAD: Normalized Residual Trajectory Modeling for End-to-End Autonomous Driving

Zhiyu Zheng, Shaoyu Chen, Haoran Yin, Xinbang Zhang, Jialv Zou, Xinggang Wang, Qian Zhang, Lefei Zhang

CVPR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 11 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add ranker to base model (ablation)ResAD base · NAVSIM v1 navtestPDMS: 84.9 -> 85.1No clear effect
add trajectory residual modeling (TRM) (ablation)ResAD base + Ranker · NAVSIM v1 navtestPDMS: 85.1 -> 86.3Improved
add trajectory residual modeling (TRM) (ablation)ResAD base + Ranker · NAVSIM v1 navtestEP: 77.8 -> 80.3Improved
add point-wise residual normalization (PRNorm) (ablation)ResAD with TRM · NAVSIM v1 navtestPDMS: 86.3 -> 87.2Improved
add inertial reference perturbation (IRP) (ablation)ResAD with TRM and PRNorm · NAVSIM v1 navtestPDMS: 87.2 -> 88.8Improved
add inertial reference perturbation (IRP) (ablation)ResAD with TRM and PRNorm · NAVSIM v1 navtestEP: 81.4 -> 83.4Improved
add TRM to TransfuserTransfuser · NAVSIM v1 navtestPDMS: 84.0 -> 85.2Improved
add PRNorm on top of TRMTransfuser · NAVSIM v1 navtestPDMS: 85.2 -> 85.6Improved
add TRM to Transfuser_DPTransfuser_DP · NAVSIM v1 navtestPDMS: 84.5 -> 85.5Improved
add PRNorm on top of TRMTransfuser_DP · NAVSIM v1 navtestPDMS: 85.5 -> 85.8Improved
add PRNorm on top of TRMTransfuser · NAVSIM v1 navtestEP: 80.0 -> 79.8Hurt
replace static vocabulary with perturbed inertial referencesDiffusionDrive · NAVSIM v1 navtestmean candidate PDMS (P_m): 60.3 -> 86.1Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy